LINGUIST List 7.585

Sat Apr 20 1996

Disc: [u] for [y] substitution

Editor for this issue: T. Daniel Seely <dseelyemunix.emich.edu>

Directory

1. Alexis Manaster Ramer, Re: 7.581, Sum: [u] for [y] substitution in English/French bilinguals
2. Patrick A Mather, Re: 7.581, Sum: [u] for [y] substitution in English/French bilinguals

Message 1: Re: 7.581, Sum: [u] for [y] substitution in English/French bilinguals

I must have missed this query or discussion earlier, but I am surprised
to find no mention of the fact that Bloomfield, somewhere in his book Lg,
argued that the "natural" substitution for IPA [y] in English would be
[i], not [u], and that the English-speaking world has developed a 
tradition of using [u] for historical reasons. I am not sure whether
he is right, but there are ways of testing such a claim, e.g., by
finding out what English speakers substitute for IPA [y] in languages
for which we do not have such a tradition (and where, crucially
the spelling is not _u_ or some modification of _u_). For example,
what do English speakers do with Finnish [y], which is spelled _y_.
Someone on this list must know.
 
Or what do English speakers do when presented with foreign words with
[y] without any spelling, i.e., o/aurally? The reason this is important
is that I recall vividly that, whenever I would demonstrate Polish
unaspirated voiceless stops to a phonetics class in the US, many if
not most of the students would "hear" these as English /b,d,g/, but
as soon as they saw the spelling, they would start "hearing" /p,t,k/.
Likewise, normally in English we use our so-called retroflex continuant
rhotic for the French uvular rhotic, but in this case I have no doubt
whatever that this is tradition, supportedby the spelling. I have
encountered native English speakers who had no exposure to the
tradition, i.e., who were not very well educated (as one would
have said) or at least not educated in what we conventionally
regards as the things that make up education, and they would
invariable "hear" and reproduce the French rhotic as [h]. 
 
So it is certainly possible that Bloomfield was right in the case
of [y] as well, but it would be nice to know and not just surmise.
What, of course, would be even more interesting is if it turned
out that different dialects of English have different "natural"
substitutions, that is, that those speakers whose so-called /u/
is actually close to IPA [y] use this sound, but that those whose
/u/ is closer to IPA [u] would use /i/, as predicted by Bloomfield.
 
Of course, it need not be a matter of all-or-nothing. It could
be that both /i/ and /u/ would be used under "natural" conditions.
 
All this would seem to be rather easy to find out if the relevant
information, which must actually exist here and there, were simply
pulled together.
 
Alexis MR

Message 2: Re: 7.581, Sum: [u] for [y] substitution in English/French bilinguals

I agree with the analysis that English learners of French often reproduce 
French [y] as [u] because the salient feature for English-speakers is 
[round], [back] being redundant in the English vocalic inventory. 
However, I have 
been observing a English/French bilingual (English L1) for some time and 
I find that, although he does correctly produce [y] as a front rounded 
vowel when the segment is a syllabic nucleus, when it is a semi-vowel (in 
words like "huit"), he uses the English semivowel [w]. Somehow, the 
correct analysis of the French vocalic system has not carried over into 
the semi-vowels (English has 2 semi-vowels while French has 3). I wonder 
whether there is a developmental sequence in L2 acquisition whereby 
the case I described might be accounted for?